791. Immunity to AAV-Mediated Gene Therapy in a Random-Bred Canine Model of Duchenne Muscular Dystrophy

Abstract

Top of pageAbstract Introduction: Duchenne muscular dystrophy (DMD) is caused by mutations in the dystrophin gene. Studies in the mdx mouse model of DMD have shown that muscle membrane integrity and function can be improved by AAV-mediated delivery of a functional dystrophin protein. To assess the potential clinical utility of treating human DMD patients with AAV-mediated gene delivery, we performed a series of direct intra-muscular injections in random-bred normal dogs and in dogs with muscular dystrophy caused by a dystrophin mutation (xmd dogs). Methods: AAV serotypes 2 and 6 carrying different promoter-transgene cassettes were produced as previously described for the murine studies and purified either on a heparin column or with a combination of heparin column and cesium chloride gradient. Direct intramuscular injections of virus (1012, 1010, or 108 viral genomes per site) in a total volume of 250 ul were performed. The injection sites were biopsied under anesthesia between 2 and 12 weeks after injection. Results: Injection of either AAV6 or AAV2 expressing CMV-b-galactosidase (b-gal) induced a strong inflammatory response containing both CD8+ and CD4+ T-lymphocytes with peak tissue destruction 4 weeks following viral injection. A similar robust immune response was seen with injection of AAV6-RSV-alkaline-phosphatase, AAV6-CMV-canine-factor-IX (cFIX) and with empty AAV6 capsid alone. Additional purification of the AAV6-CMV-cFIX by centrifugation through a cesium chloride gradient did not diminish the immune response. Continuous immunosuppression with cyclosporine (CSP) and mycophenolate mofetil (MMF) largely prevented the immune response to AAV6-CMV-b-gal in a normal dog for up to four weeks and permitted robust transgene expression. The same immunosuppressive regimen did not prevent an immune response to AAV6-CMV-cFIX or AAV6-CMV-human-micro-dystrophin in an xmd dog, suggesting that a more aggressive immunosuppressive regimen might be necessary in the xmd dog. Conclusions: Taken together, our results suggest that AAV capsid proteins, or proteins associated with the capsid, elicit significant immune responses when directly injected into skeletal muscle of normal random-bred dogs. The combination of CSP and MMF effectively prevents the immune responses in a normal dog, but not in an xmd dog, possibly due to the pre-existing inflammatory nature of the DMD muscle disease or due to genetic variation in a random-bred animal population. The robust immune responses to AAV2 and AAV6 in random-bred dogs contrast with the lack of an immune response in studies by others in inbred mice and dogs. Further studies will be necessary to determine the nature of the immune responses and to develop appropriate immunosuppressive regimen for AAV gene delivery to xmd muscle.